Pulse or digital communications – Pulse code modulation – Length coding
Reexamination Certificate
1999-05-21
2003-03-11
Vo, Don N. (Department: 2631)
Pulse or digital communications
Pulse code modulation
Length coding
C375S220000, C375S264000
Reexamination Certificate
active
06532267
ABSTRACT:
BACKGROUND OF THE INVENTION
Communications systems are frequently required to support varying data rates, and can do so using a variety of variable data rate transmission techniques. In many communications channels, intersymbol interference (ISI) is present and results in degradation of the received signal.
Almost all communications channels are noisy, or have loss such that the received signal has a low signal to noise ratio (SNR). One mechanism for dealing with noise in the communications channel is to use error control coding techniques in conjunction with higher order modulation formats to increase the throughput of the channel without increasing the power of the transmitted signal. Varying the transmission rate can require changes in the overall coding used.
One method for dealing with ISI is to use a decision feedback equalizer (DFE) which uses past decisions from a slicer to estimate an ISI component which can be subtracted from the signal. However, such methods suffer from the fact that errors in the detection process will propagate in the feedback process. This problem is especially acute in coded systems since the effects of errors will have a detrimental effect on the soft decision information of the receiver. In addition to this problem, since the detector is within a feedback loop, in a coded system the redundancy of the Forward Error Control (FEC) cannot be effectively exploited to improve the performance of the DFE.
One technique for dealing with ISI and for overcoming the above mentioned problems is to use Tomlinson/Harashima-Miyakawa preceding, which in effect moves the decision feedback equalizer function to the transmitter.
However, the techniques described by Tomlinson/Harashima-Miyakawa do not directly apply to variable rate systems since the transmitted signal constellation is fixed.
For the foregoing reasons, there is a need for a preceding system allows rate control while maintaining the desirable properties of Tomlinson/Harashima-Miyakawa preceding, as well supporting the use of other preceding techniques.
SUMMARY OF THE INVENTION
The present invention provides a method and apparatus for producing a variable rate precoded signal in which a variable rate mapper receives a rate control signal and a data signal and generates a sequence of constellation size representative values based on the rate control signal, in addition to producing a series of symbols representing the input data. The constellation size representative values are transmitted to a precoder, and the symbols are mapped to a modulation symbol to produce a modulated sequence of symbols. The precoder makes use of the sequence of constellation size representative values in conjunction with the modulated sequence of symbols to produce a variable rate precoded signal.
The precoder can be a Tomlinson/Harashima-Miyakawa precoder or other type of preceding circuit. A quantizer may be used as part of the preceding process and constrains the value of the output.
The invention can be used to map data blocks onto constellations of varying size by using the rate control signal and the data signal in the variable rate mapper to generate a sequence of symbols in which each symbol value is bounded by a corresponding value in a constellation size representative signal which is generated from the rate control signal.
In a preferred embodiment, the constellation size representative values are equal to a schedule signal minus one, where the schedule signal determines the range of a symbol.
The present invention can also be used to generate a variable rate modulation encoded signal. Bits from a data signal are divided into a first block of bits and a second block of bits, with the first block of bits passing through a variable rate mapper which generates a sequence of symbols and a sequence of constellation size representative values, and the second block of bits passing through an encoder which generates an encoded block of bits.
The two blocks of bits are used in a mapping process in which each composite set of bits is mapped to a constellation point. In a preferred embodiment, the bits from the encoder conceptually represent the least significant bits of the signal to be transmitted, while the bits from the variable rate mapper conceptually represent the most significant bits of the signal to be transmitted.
An advantage of the present invention is that it can be utilized to map data blocks onto constellations of varying size, as well as to support the use of Tomlinson/Harashima-Miyakawa preceding in a variable data rate system.
REFERENCES:
patent: 5265127 (1993-11-01), Betts et al.
patent: 5881108 (1999-03-01), Herzberg et al.
patent: 5923651 (1999-07-01), Struhsaker
patent: 6101223 (2000-08-01), Betts
patent: 6115415 (2000-09-01), Goldstein et al.
patent: 6167082 (2000-12-01), Ling et al.
patent: 6272108 (2001-08-01), Chapman
patent: 6307893 (2001-10-01), Bremer et al.
patent: 6310909 (2001-10-01), Jones
Alantro Communications, Inc.
Brady III W. James
Hernandez Pedro P.
Nguyen Dung X
Vo Don N.
LandOfFree
Variable rate constellation precoding does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Variable rate constellation precoding, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Variable rate constellation precoding will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3001979